FORMULAS

advertisement
ELEMENTARY STATISTICS, 5/E
Neil A. Weiss
FORMULAS
NOTATION The following notation is used on this card:
n sample size
σ population stdev
x sample mean
d paired difference
s sample stdev
p̂ sample proportion
p population proportion
Qj j th quartile
N population size
O observed frequency
µ population mean
E expected frequency
CHAPTER 5
Probability and Random Variables
• Probability for equally likely outcomes:
P (E) f
,
N
where f denotes the number of ways event E can occur and
N denotes the total number of outcomes possible.
• Special addition rule:
P (A or B or C or · · · ) P (A) + P (B) + P (C) + · · ·
CHAPTER 3
Descriptive Measures
(A, B, C, . . . mutually exclusive)
x
• Sample mean: x n
• Complementation rule: P (E) 1 − P (not E)
• Range: Range Max − Min
• General addition rule: P (A or B) P (A) + P (B) − P (A & B)
• Mean of a discrete random variable X: µ xP (X x)
• Sample standard deviation:
(x − x)2
s
n−1
s
or
x 2 − (x)2 /n
n−1
• Interquartile range: IQR Q3 − Q1
• Lower limit Q1 − 1.5 · IQR,
Upper limit Q3 + 1.5 · IQR
• Population mean (mean of a variable): µ x
N
• Standard deviation of a discrete random variable X:
σ (x − µ)2 P (X x)
or
σ x 2 P (X x) − µ2
• Factorial: k! k(k − 1) · · · 2 · 1
n
n!
• Binomial coefficient:
x
x! (n − x)!
• Binomial probability formula:
n x
p (1 − p)n−x ,
x
• Population standard deviation (standard deviation of a variable):
(x − µ)2
x 2
σ or
σ − µ2
N
N
x−µ
• Standardized variable: z σ
• Mean of a binomial random variable: µ np
CHAPTER 4
• Standard deviation of a binomial random variable: σ Descriptive Methods in Regression and Correlation
• Sxx , Sxy , and Syy :
Sxy (x − x)(y − y) xy − (x)(y)/n
Syy (y − y)2 y 2 − (y)2 /n
and
b0 1
(y − b1 x) y − b1 x
n
• Total sum of squares: SST (y − y)2 Syy
2
• Regression sum of squares: SSR (ŷ − y)2 Sxy
/Sxx
2
• Error sum of squares: SSE (y − ŷ)2 Syy − Sxy
/Sxx
• Coefficient of determination: r 2 SSR
SST
r
(x − x)(y − y)
sx sy
√
• Standard deviation of the variable x: σx σ/ n
Confidence Intervals for One Population Mean
• Standardized version of the variable x:
z
n
or
x−µ
√
σ/ n
• z-interval for µ (σ known, normal population or large sample):
σ
x ± zα/2 · √
n
• Sample size for estimating µ:
• Linear correlation coefficient:
1
n−1
The Sampling Distribution of the Sample Mean
σ
• Margin of error for the estimate of µ: E zα/2 · √
n
• Regression identity: SST SSR + SSE
Sxy
r Sxx Syy
np(1 − p)
• Mean of the variable x: µx µ
CHAPTER 8
• Regression equation: ŷ b0 + b1 x, where
Sxy
Sxx
where n denotes the number of trials and p denotes the success
probability.
CHAPTER 7
Sxx (x − x)2 x 2 − (x)2 /n
b1 P (X x) zα/2 · σ
E
2
rounded up to the nearest whole number.
,
ELEMENTARY STATISTICS, 5/E
Neil A. Weiss
FORMULAS
• Studentized version of the variable x:
t
x−µ
√
s/ n
• t-interval for µ (σ unknown, normal population or large sample):
s
x ± tα/2 · √
n
with df n − 1.
CHAPTER 9
z
x − µ0
√
σ/ n
• t-test statistic for H0 : µ µ0 (σ unknown, normal population or
large sample):
t
x − µ0
√
s/ n
with df n − 1.
• Paired t-test statistic for H0 : µ1 µ2 (paired sample, and normal
differences or large sample):
d
√
sd / n
t
with df n − 1.
• Paired t-interval for µ1 − µ2 (paired sample, and normal differences
or large sample):
sd
d ± tα/2 · √
n
with df n − 1.
CHAPTER 11
Inferences for Population Proportions
• Sample proportion:
Inferences for Two Population Means
• Pooled sample standard deviation:
(n1 − 1)s12 + (n2 − 1)s22
sp n1 + n2 − 2
• Pooled t-test statistic for H0 : µ1 µ2 (independent samples, normal
populations or large samples, and equal population standard
deviations):
t
x1 − x2
√
sp (1/n1 ) + (1/n2 )
with df n1 + n2 − 2.
• Pooled t-interval for µ1 − µ2 (independent samples, normal
populations or large samples, and equal population standard
deviations):
(x 1 − x 2 ) ± tα/2 · sp (1/n1 ) + (1/n2 )
with df n1 + n2 − 2.
p̂ x
,
n
where x denotes the number of members in the sample that have the
specified attribute.
• One-sample z-interval for p:
p̂ ± zα/2 ·
p̂(1 − p̂)/n
(Assumption: both x and n − x are 5 or greater)
• Margin of error for the estimate of p:
E zα/2 · p̂(1 − p̂)/n
• Sample size for estimating p:
zα/2 2
n 0.25
or
E
n p̂g (1 − p̂g )
zα/2
E
2
rounded up to the nearest whole number (g “educated guess”)
• One-sample z-test statistic for H0 : p p0 :
• Degrees of freedom for nonpooled-t procedures:
2
2
s /n1 + s22 /n2
1 2
2
2 ,
s2 /n2
s12 /n1
+
n1 − 1
n2 − 1
rounded down to the nearest integer.
• Nonpooled t-test statistic for H0 : µ1 µ2 (independent samples,
and normal populations or large samples):
x1 − x2
t (s12 /n1 ) + (s22 /n2 )
with df .
with df .
Hypothesis Tests for One Population Mean
• z-test statistic for H0 : µ µ0 (σ known, normal population or large
sample):
CHAPTER 10
• Nonpooled t-interval for µ1 − µ2 (independent samples, and normal
populations or large samples):
(x 1 − x 2 ) ± tα/2 · (s12 /n1 ) + (s22 /n2 )
p̂ − p0
z √
p0 (1 − p0 )/n
(Assumption: both np0 and n(1 − p0 ) are 5 or greater)
• Pooled sample proportion: p̂p x1 + x2
n1 + n2
• Two-sample z-test statistic for H0 : p1 p2 :
p̂1 − p̂2
z √
p̂p (1 − p̂p ) (1/n1 ) + (1/n2 )
(Assumptions: independent samples; x1 , n1 − x1 , x2 , n2 − x2 are all 5
or greater)
ELEMENTARY STATISTICS, 5/E
Neil A. Weiss
FORMULAS
• Two-sample z-interval for p1 − p2 :
(p̂1 − p̂2 ) ± zα/2 · p̂1 (1 − p̂1 )/n1 + p̂2 (1 − p̂2 )/n2
• One-way ANOVA identity: SST SSTR + SSE
• Computing formulas for sums of squares in one-way ANOVA:
(Assumptions: independent samples; x1 , n1 − x1 , x2 , n2 − x2 are all 5
or greater)
SST x 2 − (x)2 /n
SSTR (Tj2 /nj ) − (x)2 /n
• Margin of error for the estimate of p1 − p2 :
E zα/2 · p̂1 (1 − p̂1 )/n1 + p̂2 (1 − p̂2 )/n2
• Sample size for estimating p1 − p2 :
• Mean squares in one-way ANOVA:
n1 n2 0.5
or
SSE SST − SSTR
zα/2
E
2
MSTR z
n1 n2 p̂1g (1 − p̂1g ) + p̂2g (1 − p̂2g )
2
α/2
Chi-Square Procedures
• Expected frequencies for a chi-square goodness-of-fit test:
E np
• Test statistic for a chi-square goodness-of-fit test:
χ 2 (O − E)2 /E
with df k − 1, where k is the number of possible values for the
variable under consideration.
• Expected frequencies for a chi-square independence test:
R·C
n
where R row total and C column total.
E
F MSTR
MSE
CHAPTER 14
Inferential Methods in Regression and Correlation
• Population regression equation: y β0 + β1 x
SSE
• Standard error of the estimate: se n−2
• Test statistic for H0 : β1 0:
t
b1
√
se / Sxx
with df n − 2.
• Confidence interval for β1 :
se
b1 ± tα/2 · √
Sxx
with df n − 2.
χ 2 (O − E)2 /E
with df (r − 1)(c − 1), where r and c are the number of possible
values for the two variables under consideration.
Analysis of Variance (ANOVA)
• Notation in one-way ANOVA:
• Confidence interval for the conditional mean of the response variable
corresponding to xp :
(xp − x/n)2
1
ŷp ± tα/2 · se
+
n
Sxx
with df n − 2.
k number of populations
n total number of observations
x mean of all n observations
nj size of sample from Population j
x j mean of sample from Population j
sj2 variance of sample from Population j
Tj sum of sample data from Population j
• Defining formulas for sums of squares in one-way ANOVA:
• Prediction interval for an observed value of the response variable
corresponding to xp :
(xp − x/n)2
1
ŷp ± tα/2 · se 1 +
+
n
Sxx
with df n − 2.
• Test statistic for H0 : ρ 0:
t SST (x − x)2
SSTR nj (x j − x)
2
SSE (nj − 1)sj2
SSE
n−k
with df (k − 1, n − k).
• Test statistic for a chi-square independence test:
CHAPTER 13
MSE • Test statistic for one-way ANOVA (independent samples, normal
populations, and equal population standard deviations):
E
rounded up to the nearest whole number (g “educated guess”)
CHAPTER 12
SSTR
,
k−1
with df n − 2.
r
1 − r2
n−2
Download